Liang Zhu1, Meng-Hua Dai2, Shi-Tian Wang1, Zheng-Yu Jin3, Qiang Wang4, Timm Denecke5, Bernd Hamm5, Hua-Dan Xue1. 1. Peking Union Medical College Hospital, Department of Radiology, Shuaifuyuan No.1, Dongcheng District, Beijing, 100730, China. 2. Department of General Surgery, Peking Union Medical College Hospital, Shuaifuyuan No.1, Dongcheng District, Beijing, China. 3. Peking Union Medical College Hospital, Department of Radiology, Shuaifuyuan No.1, Dongcheng District, Beijing, 100730, China. Electronic address: jin_zhengyu@163.com. 4. Department of Gastroenterology, Peking Union Medical College Hospital, Shuaifuyuan No.1, Dongcheng District, Beijing, China. 5. Department of Radiology, Charité- Universitätsmedizin Berlin, Charitéplatz 1, Berlin, Germany.
Abstract
OBJECTIVE: To determine the prevalence of multiple solid pancreatic lesions on dynamic enhanced CT performed for suspected pancreatic diseases, and to identify CT features of non-malignancies. METHODS: We investigated 8096 consecutive patients who underwent dynamic enhanced CT pancreas protocol at a tertiary center over 40 months. The final clinical /pathological diagnosis served as reference standard. The diagnostic accuracy of dynamic enhanced CT for non-malignancies was calculated. A univariate and multivariate analysis was conducted to identify features that predict non-malignancies. RESULTS: Multiple solid pancreatic lesions were identified in 121 patients. The prevalence of non-malignancies was 19.8% (24/121). The most common non-malignancy was autoimmune pancreatitis (n = 21; 17.4%). Common lesions with malignant potential included neuroendocrine neoplasia (n = 62; 51.2%), ductal adenocarcinoma (n = 15; 12.4%), metastasis (n = 9; 7.4%), and lymphoma (n = 7; 5.8%). Dynamic enhanced CT had a sensitivity of 79.2% and a specificity of 92.8% for diagnosing non-malignancies. Elevated serum IgG4 level (p < 0.001), hypo-enhancement in arterial phase (p = 0.001), hyper-enhancement in equilibrium phase (p = 0.009) and location in both proximal and distal pancreas (p = 0.036) were predictors of non-malignancies, whereas pancreatic duct morphology and vascular invasion status were not. CONCLUSION: Multiple solid pancreatic lesions were rare, with a wide spectrum. Dynamic enhanced CT provides clues for identifying non-malignancies.
OBJECTIVE: To determine the prevalence of multiple solid pancreatic lesions on dynamic enhanced CT performed for suspected pancreatic diseases, and to identify CT features of non-malignancies. METHODS: We investigated 8096 consecutive patients who underwent dynamic enhanced CT pancreas protocol at a tertiary center over 40 months. The final clinical /pathological diagnosis served as reference standard. The diagnostic accuracy of dynamic enhanced CT for non-malignancies was calculated. A univariate and multivariate analysis was conducted to identify features that predict non-malignancies. RESULTS: Multiple solid pancreatic lesions were identified in 121 patients. The prevalence of non-malignancies was 19.8% (24/121). The most common non-malignancy was autoimmune pancreatitis (n = 21; 17.4%). Common lesions with malignant potential included neuroendocrine neoplasia (n = 62; 51.2%), ductal adenocarcinoma (n = 15; 12.4%), metastasis (n = 9; 7.4%), and lymphoma (n = 7; 5.8%). Dynamic enhanced CT had a sensitivity of 79.2% and a specificity of 92.8% for diagnosing non-malignancies. Elevated serum IgG4 level (p < 0.001), hypo-enhancement in arterial phase (p = 0.001), hyper-enhancement in equilibrium phase (p = 0.009) and location in both proximal and distal pancreas (p = 0.036) were predictors of non-malignancies, whereas pancreatic duct morphology and vascular invasion status were not. CONCLUSION: Multiple solid pancreatic lesions were rare, with a wide spectrum. Dynamic enhanced CT provides clues for identifying non-malignancies.